Predicted neoantigens can be relevant to the clinical outcome only when their affinity to MHC I molecule is significantly higher than the corresponding wild type peptide, and when they do not show any homology to unrelated self antigens (e.g. true predicted neoantigens - TPNAs). In addition, they may show homology to pathogen-derived antigens, representing a “super” neoantigen if the patient is immunologically primed. A novel algorithm for prediction of TPNAs presented by each patient's autologous HLA molecules was applied to datasets of all 377 HCC patients available at The Cancer Genome Atlas (TCGA). The number of neoantigens was highly variable between different HCC patients with an average of 14.5 per sample. However, considering those with the highest predicted antigenic quality, the average of neoantigens per samples with such parameters dropped down to 1.5. Results showed that patients’ survival did not correlate with the quantity nor the quality of neoantigens. However, only when the cytotoxic activity of tumor infiltrating lymphocytes was low, a correlation between neoantigens and survival was observed, suggesting that in this setting a higher number of neoantigens is required for a more pronounced efficacy on the patients’ survival. Predicted neoantigens with homology to unrelated self antigens or pathogen-derived antigens were identified. Nevertheless, a single neoantigen per patient with such a homology did not influence patients’ survival. The same algorithm was applied to a cohort of hepatocellular carcinoma (HCC) patients enrolled at our Institute, confirming the results obtained on the TCGA cohort. Interestingly, TPNAs with highest homology to pathogen’s antigens were found in the only HCC long-term survival patient. A pre-existing T cell immunity specific for these TPNAs was revealed in the patient, possibly explaining the favourable clinical outcome. In conclusion, the new algorithm allowed the identification of the very few TPNAs in HCC samples. The relevance of quantity and quality of neoantigens in the evolution of HCC without treatment with checkpoint inhibitors (CI) appears to be less central, unless the tumor cytotoxic lymphocyte infiltration is low. The identification in a single patient, of a primed immunity against a neoantigen homologous to a pathogen-derived antigen, can possibly explain the long term survival. Therefore, the identification of high quality neoantigens remains a priority for developing personalized active immunotherapies also in the HCC model characterized by a low mutational burden.

Note: This abstract was not presented at the meeting.

Citation Format: Maria Tagliamonte, Angela Mauriello, Roberta Zeuli, Annacarmen Petrizzo, Maria Lina Tornesello, Michele Ceccarelli, Franco M. Buonaguro, Luigi Buonaguro. Neoantigen load, tumor immune infiltration and prediction of survival in HCC patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 1198.